Disseminated disease is the primary cause of death in breast cancer and identifying molecular mechanism(s) that control formation and growth of metastases is critical for developing life-saving treatments. Metastasis is a multi-step process that often involves transitions of tumor cells between epithelial and mesenchymal states. This plasticity has been observed in breast cancer cell models in vitro and in tumors. In addition, breast cancers that express a mesenchymal gene signature are highly metastatic. Thus, discerning mechanism(s) that control epithelial/mesenchymal properties and epithelial to mesenchymal transition (EMT) should provide significant insights into the processes that contribute to metastatic disease. Our studies revealed that KLF4 (Krppel-like Factor 4) is a dominant regulator of the epithelial/mesenchymal status in human breast epithelial cells. Sustained expression of KLF4 is necessary to maintain an epithelial phenotype in non-transformed cells, while KLF4 overexpression induces a switch from a mesenchymal to epithelial state in breast cancer cells. KLF4 stimulates transcription of the E-cadherin gene, an epithelial determinant, and inhibits expression of Snail, an inducer of the mesenchymal phenotype. These data indicate that KLF4 is a hub for regulating the epithelial/mesenchymal states of breast epithelia. Supporting this possibility, EMT induced by Transforming Growth Factor- (TGF) is accompanied by a repression of KLF4 expression. The ability of KLF4 to enforce the epithelial phenotype is contrasted by KLF8 which drives a mesenchymal phenotype and whose expression is stimulated by TGF. Hence, Krppel-like factors determine the epithelial or mesenchymal states of breast epithelial cells and may underlie metastatic capacity. This proposal focuses on delineating mechanism(s) of action of KLF4 in dictating the epithelial state of breast cells. Since KLF4 is repressed during TGF-induced EMT, we will first determine if KLF4 inhibits TGF-induced EMT and if this involves KLF4 downregulation of Snail. This will reveal whether the precise level of KLF4 is critical for determining the extent of EMT induced by TGF and possibly, metastasis. Secondly, we will assess whether KLF8 and KLF4 are antagonistic within the same cells. If so, this would indicate that the relative concentrations of these two proteins defines the epithelial/mesenchymal status of breast epithelial cells and this possibility will be further assessed by examining the relative expression of KLF4 and KLF8 in human tumors and their correlation with E-cadherin expression. In the third aim, we will assess the impact of KLF4 overexpression on metastasis using an experimental metastasis model as well as a transgenic model of autochthonous breast cancer. Lastly, we expect that KLF4 maintains epithelial characteristics by regulating transcription of a spectrum of genes. We will use genome-based approaches to identify these genes and assess their contributions to the epithelial phenotype. Together, these studies will reveal the molecular circuitry controlled by KLF4 that modulates the epithelial phenotype and has significant implications for specifying metastatic potential.